Governing Through Scalar Elasticity: An Analysis of the Accountability Gap in Migration Control in the Central Mediterranean

Introduction

In recent years, the maritime border in the Central Mediterranean has attracted considerable scholarship that highlights the deadly consequences for migrants of border controls (Albahari, 2015; Dickson, 2021; Mann, 2018; Pécoud, 2020). The space is marked by refugee's shrinking access to legal asylum procedures and ever-expanding extraterritorial migration control measures. In a deterrence paradigm designed to keep migrants and refugees from ever-reaching jurisdictions in the global North, the accountability gap in migration control is raising more and more concern from scholars and practitioners alike (Costello and Mann, 2020). These discussions are dominated by legal scholars who examine the accountability gap by analysing international EU and domestic law, and integrating various geographical perspectives on human rights violations and their available remedies (ibid.). The Central Mediterranean has become a kind of legal battleground criss-crossed by litigation efforts to ‘overcome’ (Pijnenburg and van der Pas, 2022: 401) accountability challenges. Nonetheless, sufficient attention has yet to be paid to how accountability gaps are produced, beyond the observation that political responses to legal accountability are often problematic (Mann, 2013) and new approaches are needed (Tan and Gammeltoft-Hansen, 2020). In this article, I delve deeper into the governance of the Central Mediterranean, analysing dynamic judicial decisions on land that mould practices of executive force at sea and beyond. I argue that maritime borders are characterised by elastic scales of governing; because state, territory, law, and politics cannot be neatly hierarchised or mapped onto one another (Basaran, 2020), accountability is complicated and elusive.

My article scrutinises practices of strategic litigation ¹ that have sought to destabilise European externalisation efforts, especially those related to the regime of cooperative interdiction that emerged post-2017, after the Libya–Italy Memorandum of Understanding. ² By studying the movement between the legal reasoning of lawyers demanding accountability for border violence and the evolution of externalisation policies themselves, we gain a deeper, layered understanding of how the ever more restrictive migration control and deterrence measures in the Central Mediterranean ‘operate’. Litigation seeking accountability for violations related to border violence provides a window on the dynamic undertakings of sovereignty, which is continuously ‘(re)constituted in the unfolding, unstable pragmatics of the present’ (Richland, 2013: 213–214).

I start by showing how scalar visions of the EU's maritime border, ‘which provide for a neat separation and hierarchy of scales’ (Basaran, 2020: 23) have been complicated and blurred by the creation of the Libyan Search and Rescue Region (SRR) in the Central Mediterranean. This space is governed by different kinds of authority that cannot be easily mapped onto a central, territorially defined form of state sovereignty. Then, I outline how scalar blurring complicated the legal work of the Sciabaca&Oruka Project, a strategic litigation project that was part of the Italian research, litigation and advocacy association ‘ASGI – Associazione per gli Studi Giuridici sull'Immigrazione’, which focuses on monitoring and defending the fundamental rights of migrants and refugees. To address the maritime border's scalar elasticity, lawyers had to constantly shift their legal reasoning as they attempted to cover accountability gaps created by the indirect forms of control migrants were subjected to during forcible returns at sea. Analysing these practices unveils the structures the lawyers sought to destabilise. A dynamic picture emerges of the legal architecture of the interdiction regime ³ and the related mechanisms that continually produce new forms of responsibility diffusion for border-related violence and death at sea.

My approach in this article combines a conceptual analysis of how governing takes place across different scales with an examination of the litigation that attempts to render these harmful practices accountable. This enables me to show how bordering practices occurring across elastic scales of governing are inherently related to legal boundaries that are being made and remade. Following an analytical approach that advocates for ‘a multiplicity of groundings’ (Kahn, 2017: 19) across scales reveals how contemporary forms of executive power are inherently linked to the possibility of reasserting statist visions ‘from anywhere’, without being constrained to govern on the same scale (Basaran, 2020: 31).

Methods and Fieldwork

The data discussed in this article is taken from fieldwork conducted as part of my doctoral research on governance of the EU's external maritime border in the Central Mediterranean. I conducted eighteen months of fieldwork between February 2018 and September 2020 in various locations relating to the Central Mediterranean border zone. This included six weeks of participant observation on an non-governmental organisation (NGO) rescue vessel in the newly declared Libyan SRR. To understand the genealogy of this new rescue region under Libyan coordination, I conducted thirty-eight semi-structured interviews of international staff members, EU officials and national representatives involved in Integrated Border Management (IBM) efforts in Libya as project managers, policy advisors or representatives of donor states. The interviews were carried out on various field trips to Tunis and Brussels in 2018 and 2019. This research provides the background and context for what is discussed in this article.

I also draw from nine months of fieldwork following the work of the ‘Sciabaca&Oruka’ ⁴ Project in Rome. ‘Sciabaca&Oruka’ was set up to address and document the externalisation regime's encroachment on migrants' rights in and around the Central Mediterranean. Engaging in participant observation with project team members, I focused on the various practices needed to gather information to build profiles of responsibility and reconstruct scenes of situations of violence (Merry, 2017). Here, I follow Adler and Pouliot (2011: 5) who outline ‘practices’ in international arenas as ‘not merely descriptive “arrows” that connect structure to agency and back’, but see them as ‘dynamic material and ideational processes that enable structures to be stable or to be evolved, and agents to reproduce or transform structures’. During my fieldwork, I examined seven legal challenges addressing national and regional courts across several levels of jurisdiction, including civil and administrative courts in Italy, the European Court of Auditors (ECA), the European Court of Human Rights (ECtHR) and the UN Human Rights Committee. I analysed court proceedings and followed discussions with Sciabaca&Oruka Project members as they decided what legal avenues to pursue. Additionally, I followed the process of filing and submitting Freedom of Information (FOI) requests to various Italian ministries and European agencies and authorities to gather information on implementation of IBM programmes relating to strengthening Libya's terrestrial and maritime borders. IBM is an integral part of the EU's techniques and strategy relating to the externalisation of border control ⁵ and was first introduced by the EU Commission in 2002. It is a key policy objective inscribed in the Lisbon Treaty (Moreno-Lax, 2011), encompassing central components such as the Schengen Borders Code and operational cooperation between Member States (including Frontex; ibid.) and including various elements, such as the detachment of liaison officers from EU member states to third states of migration origin and transit, information exchange, readmission agreements, and transfer of knowledge and border management techniques through the training of officials in third states (Le Chêne, 2017: 121).

I also conducted ten semi-structured interviews with Sciabaca&Oruka and ASGI lawyers and project officers, which were supplemented with many other informal conversations. I coded the court proceedings, FOI documents, and interview transcripts using the qualitative data management software ATLAS.ti, building analytical categories from themes emerging from the data.

Statist Visions and Transnational Practices of Migration Control

Cartographies of executive power usually derive law and politics from territory; it is territory that renders power ‘visible and legible’ (Basaran, 2020: 25). Such visions of power are statist because they inscribe state authority as the central locus of politics within territory (ibid.: 26) and are particularly dependent on the idea of territorial jurisdiction. The territorially defined state is also the central building block of international law, with the principle of sovereign equality inscribed in Article 2 of the Charter of the United Nations. The statist scale, where the state is a unit that can be scaled up or down (with the division of local, national and international), orders the world and has been naturalised to enable and authorise ‘modern techniques of governing’ (Basaran, 2020: 26). This vision is what Jeffrey Kahn (2018: 11) would call a cosmology: a way of organising and imagining the way things and persons are ordered in relation to state authority that is neither complete nor fully stable. In fact, scholarship that examines practices of law and deconstructs notions of power has contributed to important critiques of this scalar order. For one thing, it is nothing new to point out that many forms of state power and control are exercised outside of territorial boundaries. This raises a host of questions about the distribution of such privileges and the role of law in global assemblages of power (Margolies et al., 2019). Although this statist scalar vision has been naturalised, and authorises and legitimates various techniques of governing in much of the contemporary world, it is obvious that these scales cannot be neatly distinguished or hierarchised in many situations where sovereign power is reconfigured or obscured. This is especially true when law operates outside of territory or ‘beyond the state’ (Eckert and Dann, 2020).

Contemporary migration governance at the external borders of Europe has been described as ‘multileveled’ (Lavenex, 2016). Complex arrangements often underpin migration policies that involve the cooperation of different states and actors spanning several jurisdictions, or have an extraterritorial character – thus, challenging assumptions of territorial jurisdiction. These characteristics are especially apparent in migration control and externalisation policies designed to ‘manage’ and ‘protect’ the borders of the global North. Most of the time, these complexities are firmly inscribed in the deterrence paradigm (Gammeltoft-Hansen, 2014; Gammeltoft-Hansen and Feith Tan, 2016), which seeks to limit migrant access to protection procedures. These policies not only hinder access to asylum in physical and legal ways, but raise accountability issues for migrants who are the target of these policies (Costello and Mann, 2020) and fall victim to border-related violence.

Moreover, the growing number of actors involved in deterrence efforts further complicates the attribution of responsibility. In the Central Mediterranean, these range from military actors, coast guards, international organisations, the governments of European states and Libya, and armed groups. Drawing on insights from legal geography, Tan and Gammeltoft-Hansen (2020: 338–339) proposed adopting the perspective of the site where violations took place in the context of migration control, instead of departing from legal texts to address systemic lack of accountability. This would map different legal approaches to establish accountability from the perspective of the site. However, a socio-legal perspective on how gaps in accountability are produced in the first place has not been thoroughly examined before.

Accountability within political systems that profess commitment to the Rule of Law, is intimately linked to the idea that the law has the capacity to hold powerful actors to account (Berger and Lake, 2018). ‘Accountability’ carries a slightly different legal connotation than ‘responsibility’, and relates to time differentially. Responsibility denotes an ongoing duty or obligation, whereas accountability requires justifying actions or decisions. Thus, to understand the production of accountability gaps, we must examine the issue of jurisdiction as a determination of ‘the power to speak the law’ (Kahn, 2018: 11; see also Richland, 2013; Schiff Berman, 2020: 17). To do so, I look pragmatically at how authoritative readings of law link to jurisdiction, following conceptions of sovereignty that move away from political ontology ⁶ debates about the nature of sovereignty (Basaran, 2020; Kahn, 2017, 2018). This enables me to show how contemporary forms of sovereignty dilute accountability precisely because of their ability to govern through scalar elasticity, whilst maintaining the illusion of governance based on a statist vision. To illustrate this, the following section examines how the maritime border in the Central Mediterranean has evolved as a site characterised by government through scalar elasticity.

The Central Mediterranean, a Legal Battlefield

In the Central Mediterranean, the recent empowerment of the Libyan Coast Guard (LYCG) has made it increasingly difficult to demonstrate the direct responsibility of EU actors for violations of migrants' fundamental rights. This is not only a question of extraterritorial jurisdiction: fundamental rights are embedded in EU legal architecture and apply extraterritorially in numerous scenarios (Moreno-Lax and Costello, 2014). Rather, this is due to fundamental changes in the way control is exerted on people fleeing the region by sea, which evolved since the fall of the Gaddafi regime in 2011. Since 2017, EU ‘orchestration’ (Müller and Slominski, 2020) techniques to establish the LYCG as an authority of reference for border patrol and Search and Rescue (SAR) activities have enabled the routine interception and return of migrants to Libya (Amnesty International, 2021; see UN OCHR, 2021).

In particular, a new Libyan SRR, of which the International Maritime Organisation's global SAR plan was notified in June 2018, contributed to operationalising the LYCG as a legitimate authority on the international stage. Due to the LYCG's role in the governance of migration at the EU's and Italy's external borders the LYCG's authority was never really accepted by non-governmental actors active in Mediterranean SAR. In June 2020, several NGOs and individuals called for revocation of the Libyan SRR on the basis of the incompatibility of the LYCG's duties with international human rights provisions (Statewatch, 2020). However, the creation of the Libyan SRR contributed to turning the LYCG, a contested actor on the international scene, into a legitimate partner for the EU and Italy. The creation of the Libyan SRR had important consequences for the way in which control was exercised over migratory movements in the Central Mediterranean. It added an extra layer of jurisdiction – or legal distance – between the policy enablers and those who fall victim to it (see also Moreno-Lax and Lemberg-Pedersen, 2019, on distance creation through law). This has made it possible for European authorities to carry out forcible returns by proxy. European actors never come in direct contact with migrants in distress, but inform Libyan authorities of the whereabouts of boats in need of rescue. The LYCG then intercepts the boats before they reach the more northerly SAR zones of Malta and Italy. ⁷ In effect, the Libyan SRR's new layer of jurisdiction modifies the ‘where and when’ of the power to speak the law (Kahn, 2017: 13). Migrants are indirectly subjected to EU and Italian control measures without directly entering their jurisdiction.

The notification of the Libyan SRR needs to be understood as emerging from a crucible of litigation that shaped the legal landscape of the Central Mediterranean after Italy started collaborating with Libya on maritime interceptions and border policing. The ECtHR's 2012 Grand Chamber ruling, Hirsi Jamaa & others v Italy, was a historical judgement in that regard. At the time, the court condemned Italy's instrumental role in the forced return to Libya of eleven Somali nationals and thirteen Eritrean nationals. Italy claimed that the legal basis of the collaboration between Italian and Libyan officials was the 2007 Additional Operating and Technical Protocol on cooperation in the fight against irregular immigration, the 2008 Treaty on Friendship Partnership and Cooperation and an additional ‘technico-operational’ executive protocol adjoined to the 2007 Protocol in 2009 (Giuffré, 2012: 700–701). The Hirsi decision featured the following historic points. First, the court confirmed the extraterritorial jurisdiction of States that were parties to the convention when they ‘exercised authority and control over a person outside [their] territory’ (Pijnenburg, 2018: 400). Indeed, the judgement affirmed that when the applicants were aboard an official Italian ship, they were under the ‘continuous and exclusive de jure and de facto control of the Italian authorities’. In other words, the court recognised Italian extraterritorial jurisdiction in this scenario. However, another important aspect of the decision indicated that Italy could not use international obligations arising from other international regimes it was bound to (such as the law of the sea or the bilateral agreement with Libya), to circumvent its responsibilities under the ECHR (ibid.).

In relation to Hirsi, the development of the Libyan SRR can be seen as a perfect illustration of what Mann (2013) has called the ‘dialectics of transnationalism’. He shows how, even when judicial review of interdiction policies manages to be transnational – for example as was the case for Hirsi – the opposing forces of executive and judicial power bifurcate, so that the transnational legal environment ends up instructing ‘executive agencies on how to further bifurcate executive and judicial functions, avoiding the costs imposed by human rights law’ (ibid.: 318). In her analysis of the ‘unintended consequences’ of Hirsi, Greenberg (2021: 519) calls ‘counterpedagogy’ the phenomenon of state actors learning from international human rights courts how to effectively ‘shore up the exercise of state power’. Policy and operational developments linked to the emergence of delegated pushbacks through the ‘contactless control’ (Moreno-Lax, 2020; Moreno-Lax and Giuffré, 2017) of the LYCG seem to stem from government practice adapting to the Hirsi judgement to avoid legal liability.

Establishment of the Libyan SRR and the legal screen it provides for forced returns carried out on the high seas is a tour de force of liability avoidance. Nonetheless, the violent developments in the Central Mediterranean's externalisation regime are being contested by legal and political means. Pending at the ECtHR, S.S. and others v. Italy is attempting to prove Italian jurisdiction in a delegated pushback case. It concerns the dramatic interception, by the LYCG's patrol vessel, Ras al Jadar, of 150 people on a capsizing boat. Several people died and the others were pushed back to Libya on 6 November 2017. It led the legal scholar Violeta Moreno-Lax, lead counsel on the case, to develop a new theory of jurisdiction which she calls ‘functional’. Her argument is that Italy exercised a ‘sufficient degree of “effective control” over the applicants' fate’ to reach the jurisdictional threshold of the Convention (Moreno-Lax, 2020: 387). As mentioned with respect to Hirsi, the application of extraterritorial jurisdiction followed the ‘personal’ model of jurisdiction, rather than the ‘spatial’ model informed by public international law norms (see Moreno-Lax and Costello, 2014) and pivoted around issues of whether the Italian authorities exercised ‘full and exclusive control’ over the applicants. Moreno-Lax points out that the ECtHR made clear that ‘direct physical contact is not always necessary as long as the control thereby exerted is indeed effective’ (ibid.: 401). The sophistication of the deterrence policy in the Central Mediterranean forces her to develop a new theory of jurisdiction that is not territorial but functional: according to her formulation, jurisdiction on a given case should depend on the ‘governmental “functions” through which the power of the state finds concrete expression’ (ibid.: 402). This means that ‘policy measures and operational procedures’ through which states exert ‘personal or spatial control’ (ibid.: 403) should be understood as public powers that amount to jurisdiction.

The theory of functional jurisdiction is an attempt from within doctrinal normativism (Costello and Mann, 2020: 327) to address these executive dark spots, which come from the ‘disaggregated sovereignty’ (Mann, 2013: 328) emerging in transnational migration control. To further illustrate how executive dark spots, or accountability gaps, are produced throughout the Central Mediterranean regime of externalised control, the next part of this article analyses other litigation practices that attempt to achieve accountability via various legal means. Analysis of arguments made by Sciabaca&Oruka Project lawyers sheds light on dynamic developments in bordered jurisdictions as legal contestations of border violence reveal the scalar elasticity of maritime border enforcement in the Central Mediterranean.

Litigating Against Forced Returns at Sea

Costello and Mann (2020: 313–314) outline four different, but related features of migration control that ‘render accountability elusive’: (1) the asymmetry in international law between the right to leave a given country and the right to enter, (2) the narrow legal category of refugee, which does not correspond to the contemporary range of threats people may be fleeing from, (3) the increasingly extraterritorial nature of migration control and the contested applications of state jurisdiction related to it, and (4) the fragmentation of actors involved in migration control. This enumeration hints at the issue of jurisdictional scales, but does not address how ‘scalar elasticity’ (Basaran, 2020) is manipulated strategically by state authorities within border externalisation policies. In such policies, the statist scalar order – which can be divided into the local, national, and international – equates sovereignty with territorial boundaries. In practice, these orders are geographically indeterminate and constantly redrawn and blurred by the governance of borders. By paying greater attention to the legal reasoning that goes into strategic litigation, and by tracing the outcomes of cases in which border violence is contested, we garner deeper insights into how accountability is rendered elusive by powerful scalar visions. Del Mar (2017: 44) elaborates on the possibilities of studying the relational dimension of legal reasoning that enable inquiries into norm-generating activities. Taking into consideration the role of imagination, the practice of legal reasoning across time and the relational dimension of legal reasoning contributes to ‘analyzing how legal reasoning within one norm-generating unit relates with that of other norm-generating units’. An analysis of legal reasoning, followed by tracing case outcomes across different judicial levels, gives insight into the scales of legality of the EU border regime, and how accountability gaps are produced and maintained. It also shows how unaccountability is produced by and through law, as authoritative actors strategically wield the spatial and temporal elasticity of governmental scales to render legal contestations ineffective.

As supporting evidence, this section examines litigation developed within the Sciabaca&Oruka Project to address the effects of border externalisation on fundamental rights in the Central Mediterranean. In particular, I focus on models of legal reasoning that were affected by the complex, shifting legal architecture of the border regime. Evolving legal models of responsibility attribution were shaped by the increasing difficulties lawyers faced with more conventional human rights litigation in situations related to border violence. Besides dialectics of transnationalism that render parts of the executive web of force out of legal reach, public litigation in favour of migrants having been pushed back at the EU's maritime border is complicated by practical challenges. Proving the identity of a victim and keeping in touch with that person can be difficult. Securing evidence that a violation took place and the time it takes for judgements to be rendered are also problematic. Indeed, migrants travelling across international borders may have outdated identification or may have lost documents along the way. Along their journey, they may lose touch with lawyers representing them in other countries. State officials engaged in pushback may not have identified these people before sending them back over the border.

A landmark judgement in 2019 of the civil court in Rome (Tribunale di Roma, Prima sezione civile Sentenza n. 22917/2019, RG n. 5615/2016) on the Italian navy's expulsion at sea of fourteen Eritrean citizens ten years earlier, illustrates some of these practical difficulties. ⁸ Lawyers from Sciabaca&Oruka who had worked on the civil case told me how they had struggled with challenges specific to the litigation of summary expulsions at sea. On the question of identification, the Eritreans had luckily been filmed and identified aboard the Italian ship, providing photographic evidence to show the judge. The pictures formed the crux of their case. However, the Roman court asked for further guarantees that the people in the photos were the same people transferred to the LYCG's vessel in 2009. It emerged that the United Nations High Commissioner for Refugees (UNHCR) had subsequently identified them in the Libyan prison where they had been held upon disembarkation: it had collected their identities and given them protection cards. The lawyers pointed out to me that they sometimes had difficulty getting information from international organisations like the International Organization for Migration (IOM) and UNHCR, which often carried out the identification of migrants who had been returned to Libya. Thus, fragmentation across actors involved in the deterrence regime presented them with practical problems.

In the civil case in Rome, the Italian judge had ordered the UNHCR to produce the identification documents for the purpose of the trial. However, the UNHCR raised the principle of the immunity of international organisations (see Verdirame, 2011) and refused to cooperate because it was not subordinate to Italian judicial authority (la clausola di non sottoponibilità a una autorità giudiziaria italiana in the words of one of the lawyers). The lawyers were finally able to gather other forms of evidence through testimonies of different people around the world confirming the identity of the people whose photos had been taken on the boat.

Besides the difficulty of getting in touch with people who had been pushed back, and staying in touch over long periods of time, ‘events’ that had played out (or not played out – when it came to the absence of procedures and remedies) had to be documented and the state actors involved had no incentive to do so. The lawyers needed access to this evidence in a form that could be received in a court of law. They also commented on the casual informality that increasingly characterised authorities' behaviour in migration control, especially at the external border: ‘there is a clear incentive for governments to do less, to act less accurately, to do things in a less precise way. The worse that governments do their job, the fewer instruments we have to react against them’, one of them said. For them, the shift to delegated forms of control fitted squarely within this trend. That same lawyer added:

Now all Italy has to do is call up the Libyans who will come and pick the people up. So how is a person whose rights have been violated meant to act when it never came into direct contact with the authority that notified the Libyan Coast guard to pick them up? How can I act if I was never identified by that authority?

Shifting Models of Responsibility to Address Elastic Scales of Governing

Presented with the multiplication of migration control policies and the difficulty of addressing them via direct causation models linked to violations, Sciabaca&Oruka lawyers and project collaborators started diversifying their claims against various types of harm creation and lodging these claims at domestic and regional, as well as national and international judicial levels. One key shift in their legal reasoning was to widen the perspective from direct situations of rights violations to include the structural conditions that had enabled them. The Sciabaca&Oruka Project increasingly analysed how externalisation policies functioned by tracking related funding and expenditure decisions. This involved filing many FOI requests. How public money is spent is fundamentally linked to questions of state accountability. It has additional bearing on questions of citizen accountability for government misdeeds. A key element of Sciabaca&Oruka's strategy extended beyond the harmful effects of externalisation on the rights of migrants (at sea and beyond), to highlight the ways in which citizens are materially implicated in the actions of their government (through taxes; see Rothberg, 2019).

The collective right to responsible government was articulated in claims on the Italian domestic level and on the EU level. For instance, as part of the Sciabaca&Oruka Project, a complaint was submitted to the EU court of auditors by three advocacy associations, the Global Legal Action Network (GLAN), ASGI and the Italian Recreational and Cultural Association (ARCI). The complaint centred around a controversial EUTF IBM project implemented in Libya. One of its main stated objectives was ‘to improve the Libyan capacity to control their borders and provide for lifesaving rescue at sea’ (EUTF, 2017: 2), including supporting the Libyan authorities' proposal of a new Libyan SAR zone to the International Maritime Organisation. The complaint to the ECA argued that the EU was contributing to human rights violations by funding the IBM project since it envisaged shifting the responsibility for SAR activities in the new Libyan SRR to Libyan authorities who were directly involved in human rights abuses. A legal opinion on the legality of EU funding for the LYCG stated that funding for the project, which stemmed partially from the European Development Fund (EDF), was inconsistent with the development objectives to which the EDF was legally bound. This constituted a violation of EU Parliament budgetary authority (Dann et al., 2020).

Here, I focus on another case that attempted to address the misuse of public money and was filed at the Italian domestic level. On 30 October 2017, ASGI lodged a lawsuit against the Ministry of Foreign Affairs (MAECI) disputing the so-called Fondo Africa decree (Decreto 41100/47), which provided 2.5 million euros for technical support to the Italian Ministry of Interior. These funds were to support competent Libyan authorities in improving border and migration management, stopping the smuggling of migrants and strengthening SAR activities. ASGI argued that the funding was actually used to violate the terms of the law that had established the Fondo Africa. The stated objectives of the fund were to foster cooperation, instigate a dialogue with specific countries along the migration route, fight the root causes of migration and reinforce protection measures for migrants along the route. Instead of doing this, ASGI argued, the measures financed by the fund appeared to pursue the sole objective of reducing migration flows to Italy and Europe and, in doing so, negatively impacted the fundamental rights of migrants. In January 2019, the administrative court of the Lazio region rejected ASGI's claims. ASGI then launched an appeal at the Consiglio di Stato, the highest administrative judiciary body in Italy. In June 2020, the Consiglio di Stato rejected the appeal. The court argued that, because 18 million euros of the fund had been used by the IOM for Assisted Voluntary Returns and for their work assisting migrants and local communities in Libya, and another 10 million had been given to the UNHCR for their protection work, the Italian state and its involvement in Libya via the Fondo Africa was not limited to technical assistance in border control. It also noted that it could not evaluate the usefulness or suitability of the means used to carry out the Fund's stated aims, since public administration benefitted from a margin of manoeuvre based on the ‘merit’ of its competences, which the court said it could not review. The court additionally used the argument that the Italian administration had no effective control over harmful practices carried out in Libyan detention centres. As such, Italy's intervention on the migratory route in the Central Mediterranean had to engage and cooperate with ‘the only possible interlocutor’ with direct and exclusive control over the Libyan coastline: the sovereign state of Libya. Finally, the court noted that no international court had ever condemned a cooperation intervention, like the one ASGI was contesting, for contributing to indirect refoulements or delegated pushbacks of migrants and asylum seekers.

Before the judgement on the Fondo Africa case was rendered public, one project collaborator confided to me that, even if they were to lose in front of the Consiglio di Stato, the team would do it again: to raise the issue of Italian complicity in such a way on that legal level was already an achievement. The action was notable because ‘we attack[ed] the instrument enabling harm itself’. One lawyer added that the action's main message was to signify that ‘as the Italian state you cannot give funding to a subject like Libya’. She continued:

I think it is also important that it is an association raising the problem, and therefore there are no problems of active legitimacy or need to find the victims and making sure they are safe, proving that they were on board a specific ship etc. Here, we are giving back a responsibility to the Italian government in the sense that we are confronting it and asking it to explain why it gave that money to Libya. Above all, because [the Fondo Africa] is the instrument on which everything else is based, in the sense that without patrol boats, or without those weapons, or without that money, Libya cannot do what it is doing.

In legal anthropology, Eckert (2018b: 375) has pointed out that, in our increasingly entangled world, there is a juridification of ‘claims to mutual care and concern arising from moral notions of obligation’. In particular, anthropological research has looked at how social movements and human rights defenders have, in different ways, attempted to redraw the ‘net’ of causal responsibility through legal institutions which govern attributions of responsibility in procedural, tort and corporate law (Affolter, 2020; Eckert and Knöpfel, 2020; Lindt, 2020; Loher, 2020). The corporate accountability ‘gap’, for example, has been read as a sign of disconnect between moral expectations and legal regulations within contemporary global capitalism ⁹ (Eckert and Knöpfel, 2020). The actions described above were complementary to those in which the fundamental rights of migrants were claimed directly; they mobilised administrative law and expanded perspectives on accountability. They went beyond looking at direct, causal relations of violence to consider, instead, the relational situatedness of violence, and contest the enabling conditions of situations of damage or harm creation (Eckert, 2018b). This entailed broadening the definition of who could be considered as having been ‘harmed’ by externalisation processes. The Fondo Africa case described above – came to be about the right to a transparent government, as well as the right to a government not complicit in rights violations abroad, hence, focusing on the European civilian right for accountable government. This took place through novel usage of the law (the Italian transparency decree for example) and through the development of new types of legal reasoning that stretched the notion of state responsibility in the field of migration, to include state responsibility not to harm the migrant ‘Other’, and finally to a citizen's right not to be complicit in this harm. This was an attempt, through legal argumentation, to show that there were entities in civil society (in this case, the association ASGI) that stood for something other than the abstract category of ‘people’ used by forces pushing for ever more restrictive migration control (Butler, 2015; Eckert, 2018a).

The Sciabaca&Oruka team members had to develop strategies to address the varying jurisdictional scales of border externalisation. Besides being legally complex, delegated migration control and transnational migration management blur causal chains of action and dilute responsibility due to the many different agents and kinds of agents involved (for example, in the Fondo Africa case: the Italian Ministry of Interior, the Libyan state, the IOM and the UNHCR were involved). Therefore, the lawyers' shifts in legal reasoning were intimately bound to the ways in which governance of unwanted migration at the EU's external border continually redrew the ‘inside and the outside’ of the law (Basaran, 2020: 29).

In the judgement on the Fondo Africa case, the Consiglio di Stato's arguments relied heavily on the scalar vision of territorial integrity: Italy had to cooperate with Libya as the sole authority with the legitimacy and power to enforce control over its territorial boundaries, and the Italian state had no effective control over what went on in Libyan detention centres. Still, throughout the legal proceedings, there was no denying Italy's involvement in supporting Libyan ‘management’ of migration flows.

Contrasting the reasoning of Sciabaca&Oruka lawyers to that of the Consiglio di Stato highlights how a sovereigntist, territorial scalar vision clashes with a perspective that addresses the legality of governing techniques. The latter approach is clearly dispersed across legal geographies and interconnects different scalar levels (Basaran, 2020: 28): from the international to the transnational and national levels of governance. Transnational authority extends beyond territorial boundaries and affects the plight of migrants in the Mediterranean border zone, although it often claims to be bound by principles of sovereign equality and bounded jurisdiction. By reaffirming a specific scalar application of jurisdiction to territory and reconfirming a sovereign control bound to territory, the court erected legal obstacles that further limit possibilities of seeking redress, through judicial means, for the harmful effects of transnational migration control.

This is reminiscent of Marilyn Strathern's (1996: 522) work on law as an institution that has the power to ‘cut’ into a network of social relations, to structure the possibility of claim-making by cutting into a ‘limitless expanse of “justice”’. Here the court aligns with a statist scalar vision and, by extension, ‘decouples chains of interdependencies’ (Eckert, 2016: 244) that define which relations are deemed relevant and which not. Ultimately, the combination of ‘practices of constituting scalar boundaries’ and authority disempowers ‘resistance to scalar divisions’ (Basaran, 2020: 31) such as the strategic litigation described above. Although the Consiglio di Stato judgement reaffirmed a statist vision, the practices of governing taking place at the external border clearly implicate elastic scales of governing scattered across national and international jurisdictions.

Conclusion

Accountability gaps in contemporary migration control practices are produced by authorities that declare statist scalar visions whilst, in practice, govern via elastic scales (Basaran, 2020). In this article to illustrate how this takes place across different scales of law, I have zeroed in on the practices and legal arguments made by Sciabaca&Oruka Project lawyers in Italy. Their legal attempts to contest the externalisation of control of the EU border highlight the dynamic development of jurisdictional borders. Lawyers defending the rights of migrants pushed back at sea turned towards different models of responsibility to multiply the possibilities for the EU and Italy, in this case, to be held accountable for their migration control policies. In discussing their shifts in legal reasoning, from direct to indirect models of responsibility, I have shown how the issue of jurisdiction lies at the heart of current struggles for accountability in migration control. Proving direct control (thus causation) in legal proceedings where migrants' rights have been violated in situations of externalised control has become increasingly difficult. This is due to the multiplication of actors involved in border control, but also, to the dialectics of transnationalism (Mann, 2013). The case of migration control at the EU's external maritime border is striking in that regard since competences have been increasingly transferred to ‘transnational networks beyond borders’ (ibid.: 317) which renders the judicial review of these practices elusive. Hence, litigation that addressed enabling factors of harm creation in the Central Mediterranean was an attempt to address the elastic scales of governance in maritime interdiction. Strategic litigation gives us insight into the dynamic shape that both sovereignty and migration control take on in transnational contexts of governance, and how legal scales become blurred across local, national, and international jurisdictions.

Only by looking closely at struggles for legal accountability at Europe's borders do we begin to gain a more comprehensive picture of how systematic impunity is enabled by and through law (see also Mann, 2020). Indeed, the attempts to address these trends of outsourced migration control are often outmanoeuvred by the fundamental advantage that elastic governing gives to state actors. This is particularly visible when authoritative readings of jurisdiction ‘cut’ (Strathern, 1996) through entangled webs of interrelated actors and networks involved in migrants' rights violations at the external borders of Europe. The Fondo Africa judgement reasserting the authoritative statist scalar vision of jurisdiction is a good example of such cuts in the network of governance. Accountability gaps are also caused by current mechanisms of judicial redress that privilege notions of immediate causation and intent (Eckert and Knöpfel, 2020: 2); as I showed, proving direct causation for rights violations has been rendered more difficult by the elasticity of contemporary migration governance.

Even eventual successful litigation is ultimately pyrrhic because states can redraw the lines of legal accountability, by for example, obfuscating the limits of jurisdiction as it would seem to be the case in the Central Mediterranean. The progress made by Hirsi has been for the moment circumvented by the creation of the Libyan SRR. Progressive lawyers and their litigation efforts have contributed to the shaping of executive power at the external borders of Europe over the past decade. Moreover, they continue to try to defend the right to have rights as well as use the law to counter border violence. But this ‘contrapuntal dynamic of constraining judicial intervention’ (Kahn, 2018: 57) is under threat. The elastic forms of governing that now characterise the policing of migratory movements in the Central Mediterranean and beyond are increasingly remaking the juridical landscape by turning ‘law’ into a tool of governance, whilst all the whilst eclipsing the possibilities for law to be used as a mechanism for redress (von Benda-Beckmann et al., 2009: 15).